Ayn Rands Theory of Concepts: A Brief Overview

William Thomas
Abstract: This paper summarizes Rands theory of concepts as presented
in the first five chapters of Introduction to Objectivist Epistemology.
Rands approach to universals, essentials, and definitions is discussed, and
key Randian usages such as unit, measurement, distinguishing
characteristic and Conceptual Common Denominator are explained. An
appendix summarizes Rands treatment of several concepts.
Introduction
Ayn Rands theory of concepts is presented in her monograph Introduction to Objectivist
Epistemology (a.k.a. ITOE. See the sources for a full citation; all references are to the
Second edition).
The purpose of the theory is to explain how it is, in the most basic sense, that our abstract
knowledge can be objective knowledge of reality. Knowledge is formed in theories and
propositions, but the basic building blocks are concepts, the mental content that lies behind
our use of general terms. In her foreword, Rand writes: concepts are abstractions or
universals, and everything man perceives in particular, concrete. What is the relation
between abstractions and concretes? To what precisely do concepts refer in reality? (1)
Answering these questions is the task Rand has set for herself.
Rands theory attempts to resolve a dichotomy that runs throughout the history of Western
philosophy.
On the one hand, realists, of whom Aristotle and Plato were thought to number, argued that
the referents of general terms must consist in or partake of some identical, universal nature.
Concepts therefore would have a real basis. All men would have manness, whatever that
might be, or would perhaps reflect the ideal Form of Man as existing in some dimension
known only to the intellect (as the Platonic tradition held).
On the other hand, nominalists (from Occam to the American pragmatists to Wittgenstein),
emphasized the fact that general terms are used according to human choice: no terms are
forced upon us by existence. Universals do not exist as such, and the nominalists thus tend
to hold that our concepts are not constrained by reality in any way that can be made clear.
Rand resolves this contradiction by constructing a theory that recognizes, with the
nominalists, that universals do not exist metaphysically and that human purposes and
contexts contribute to the formation of concepts. But Rands theory also insists, with the
realists, that the nature of the referents of a general term constrains the term and gives an
objective basis to the human decision to regard those referents as similar. People have no
identical man-ness to partake of, but the attributes and characteristics that people do
havetheir metaphysical identitiesprovide a real basis, independent of human thought,
for the concept of man.

Units and Concepts

Rand insists that there is no question that there is a reality which we can know. As she
adverts at the close of the foreword, for the purposes of this [work], the validity of the
senses must be taken for granted, and one must remember the axiom: Existence exists. She
begins her first chapter by stating, Consciousness, as a state of awareness, is not a passive
state, but an active process that consists of two essentials: differentiation and integration.
(5) As awareness consciousness is by its basic nature in contact with a reality beyond
itself.
Consciousness (and sense-perception as a form of consciousness) distinguishes existents in
reality by recognizing the ways in which they differ. When the eye distinguishes a bird
against the background of the blue sky, for example, that is differentiation. Consciousness
also integrates, by, for example, relating two or more existents as contrasted with other
existents. For example, when we regard two birds as a flock, because they move together,
as against the sky and against other birds that are moving independently, we have engaged
in a simple act of integration.
A concept, writes Rand, is a mental integration of two or more units which are isolated
according to a specific characteristic(s) and united by a specific definition. A word, (with
the exception of proper names) is a (visual-auditory) symbol that denotes a concept. (10)
Unit has a specific meaning in Rands theory. An existent is regarded as a unit when it is
understood as the member of a group. The referents of a concept are units because, for us
to subsume them under a concept, we must regard them as being part of the same group as
other referents of the concept, and as being distinct in this way from non-referents.
Individual cars, for example, are units of the concept of car, for all those people who
grasp that concept. The cars as such are not units of anything: units are existents as
understood and grouped by a knower.
Consider a red apple, a green apple, an orange, and a table, in a blue room. Now
perceptually, we can differentiate these existents as particulars. As such, none of these
items are units of any group. Given their characteristics, they can be obviously grouped in a
number of ways. We might differentiate the apples from all the other items. Then the apples
would be units of a group with two members. We could differentiate all the fruits from the
table. Then the apples and the orange would be units of a different group. And so on.

Measurement Omission
The basic problem of universals is how things that are not quite the same can be of the
same kind. Rands solution is to provide a refinement of the theory of similarity. Baldly,
the idea of similarity is simply that while no existents are identical in all respects, some
things are just plain similar to each other. Without an account of what similarity consists in,
it is not a satisfactory basis for a theory of objectivity.

3
Rands refinement is to define similarity in terms of relative quantitative (or in-principle
quantifiable) proximity along some dimension(s). Thus in Rands theory, similarity is not
brute or mysterious fact, it is a precise and objective relation among the characteristics of
existents. This quantifiable relation may be cardinally quantifiable, or the relative proximity
in a dimension may only be quantifiable as an ordinal ranking.
Rand gives as an example the concept of table. Tables shapes vary, she writes, but
have one characteristic in common: a flat, level surface and supports. [One] forms the
concept table by retaining that characteristic and omitting all particular measurements, not
only measurements of the shape but of all the other characteristics of tables (many of which
[one] is not aware of at the time). (12)
The shape of any given table is, in Rands terminology, a measurement. (Measurements
are, in this sense, aspects of the identity of an existent understood in the context of inprinciple quantifiability). Rand precedes her discussion of table with a discussion of a
unit of length as existing in a plainly quantifiable dimensionnamely, length or
extension. (11) In discussing table, she calls attention to the fact that any given shape is
quantifiable in terms of linear measurement in three dimensions. (14) Presumably were it
to be literally measured, shape would take a multivariate value. An existent that varied too
much in one aspect of shape, such as by having a peaked, rather than a horizontal surface,
would not be considered a table. Something with a horizontal surface and some supports,
but the size of a house, would not be considered a table (more likely, a shed). So the
measurements of shape admitted under table are constrained to be no larger than and no
smaller than some limits relevant to the purposes tables serve. Within the range of
admissible measurements, however, when we think of tables, we omit (that is, ignore for
our present purposes) the measurements. Thus when we discuss tables they can be of
glass, wood, metal, plastic, even iceat an arctic hotel in winter, with the limits that they
be solid. Tables can be of different sizes, within an appropriate range of scale. They can be
round, triangular, etc.
Rand writes that tables have one characteristic in common. This usage can be highly
misleading, because in effect she has borrowed the language of the realists without
accepting its literal meaning. In Rands theory tables all have shape in common, but then
it is true, as she would admit, that every solid physical object has some shape. What tables
have in common is that they possess measurements of shape within a certain range
(their distinguishing characteristic), but they need have no particular traits that are
literally the same nor do they partake of any literally shared identical trait.
Rand calls the dimension(s) along which we distinguish similar objects from dissimilar
ones (in any given act of concept-formation) the Conceptual Common Denominator. She
defines this neologism as the characteristic(s) reducible to a unit of measurement, by
means of which man differentiates two or more existents from other existents possessing
it. (15) In the appendix, in a discussion of the formation of a particular concept, Rand
adds: Unless you differentiate this particular grouping from another one with which it has
something in common but differs in measurement, you couldnt have a concept. Because. . .
there are two aspects of the processone is integration, but the first one is separation
[Rand appears to mean what elsewhere she calls differentiation].

4
Figure 1:
___W_________/X ___Y/_________________Z

To illustrate Rands idea, figure 1 gives a stylized representation of a Conceptual Common

Denominator. The letters represent existents that possess traits with some values upon this
CCD, which might be a shade of color, for example, or size, or intelligence, or ferocity, or
rectilinearity, or some other type of trait. X and Y are closer to each other than to W or to
Z. X and Y share the common characteristic of falling within the interval marked by the
hashes: /. . . /. One could form a concept integrating all things that fall into the interval.
Rand notes that similarity. . . is the relationship between two or more existents which
possess the same characteristic(s), but in different degree or measure. (13) For the most
part, we do not handle the complexities of measurement or determination of measurability
via conscious thought, at least not in the case of existents that we can perceive directly via
our senses. (Existents we infer from data, such as photons, are another matter: these do
require explicit measurement to some degree if only as a matter of inferring the units
traits.) Rather, Similarity is grasped perceptually; (14) our sense organs give us the
ability to directly discern and compare the identities of entities in at least some dimensions.
We can sense relative differences in temperature by touch, for example, and we can see
relative differences in length.
Similarity relations are real, but can only be measured relatively. It is true that many of our
perceptions, such as coldness or heat, for example, appear absolute to us, but this is because
our own bodies provide a contrast object against which characteristics such as size or
temperature can be contrasted. It is also worth noting again that similarity for Rand is
relative to some standard or contrast objects. No length is per-se long or short. No size
is large or small but that it is so judged relative to something else. And measurements,
as she shows with her example of length, are made using a standard of measure.
Rand insists that if two traits can be regarded as similar, this is possible because their traits
are measurable and commensurable, even if no explicit technique of measurement as yet
exists for this type of trait (she gives the example of the similarity of colors in perception,
which was only grounded in scientific measures after centuries passed. (15))
Rand returns to the concept of man in summarizing her initial presentation of her theory:
Now we can answer the question: To what precisely do we refer when we
designate three persons as men? We refer to the fact that they are living
beings who possess the same characteristic distinguishing them from all
other living species: a rational facultythough the specific measurements of
their distinguishing characteristic qua men, as well as all their other
characteristics qua living beings, are different. (As living beings of a certain
kind, they possess innumerable characteristics in common: the same shape,
the same range of size, the same facial features, the same vital organs, the
same fingerprints, etc., and all these characteristics differ only in their
measurements). (17)

In the context of the terminology of her theory, Rand elaborates her definition of concept
as follows:
A concept is a mental integration of two or more units possessing the same
distinguishing characteristic(s), with their particular measurements omitted.
(13)
(For more examples of concepts and their CCDs, see Appendix A.)
Rands use of child-development
Much of Rands exposition in ITOE is centered on an account of a childs epistemological
development, beginning even in infancy. I am not going to discuss this account or Rands
claims in any detail here. However, I would like to highlight a point of interpretation: it
seems to me that Rand has two important and valid purposes in focusing on the
development of an individuals knowledge, regardless of how exactly her theory fits certain
details of common paths of child development.
First, focusing on development as the proper context for a theory of abstraction casts the
entire discussion in a context of philosophical innocence. It avoids thereby the lamentable
tendency of philosophers to impute to nave knowers the intuitions of adults. Many thinkers
addressing these issues have taken logic and mathematics as the standards of knowledge
and certainty, despite the fact that these are mental skills that it takes years of training in
abstract thought to master.
Second, Rands accounts of child-development, while usually couched positively, in fact do
most of their work in her theory normatively. To be sure, her theory places some strict
limits on what is possible in concept-formation (to be differentiated, objects must share
some commensurable characteristic or dimension of comparison which can serve as the
CCD). But largely, her theory concerns how concepts need be properly formed, and does
not purport to be a full positive account of language as it is actually used. (T)he question
of how particular men happen to learn concepts and the question of what concepts are, are
two different issues. In considering the nature of concepts. . . we must assume a mind
capable of performing (or of retracing and checking) that process. (21)

Abstraction from abstractions

Rand argues that our most direct and full awareness of reality is via perception. Crucially
for her theory, perception is a rich form of awareness even to small children or animals
with little or no concepts nor the ability to form concepts. Thus concept-formation begins
with differentiation and integration at the perceptual level, and especially the visual level,
since vision is the dominant sense in humans. A child can learn to distinguish various items
of furniture this way, Rand says, and by contrasting them with other household objects, can
form the concept of furniture itself. (21)

6
Another good example from a childs context might be food concepts: children learn
concepts for different kinds of food: milk, juice, bread, hot-dog, etc., before they form a
clear understanding of food as a functional abstraction integrating all forms of
comestibles.
We abstract from abstractions when we use existing concepts as the basis for forming new
ones. We can do this in basically two ways, either subdividing a single concept or
integrating several conceptual categories into a broader concept.
When concepts are integrated into a wider one, the new concept includes all
the characteristics of its constituent units; but their distinguishing
characteristics are regarded as omitted measurements, and one of their
common characteristics determines the distinguishing characteristic of the
new concept: the one representing their Conceptual Common
Denominator with the existents from which they are being differentiated.
(23)
When a concept is subdivided into narrower ones, its distinguishing
characteristic is taken as their Conceptual Common Denominatorand is
given a narrower range of specified measurements or is combined with an
additional characteristic(s), to form the individual distinguishing
characteristic of the new concepts. (24)

Integrations of concept classes (such as animal, which draws together people, cats,
lizards, birds, etc.) include the entirety of the subgroups they bring together (all cats are
animals, e.g., and everything true of cats is true of some animals). (23)
Rand remarks that
Just as wider integrations of concepts require a more extensive knowledge,
so narrower subdivisions of concepts require a more intensive knowledge.
For instance, the concept father requires more knowledge than the concept
mansince it requires knowledge of man, of the act of reproduction, and
of the consequent relationship. (27)
As functional concepts, furniture and food cannot be formed simply on the basis of the
physical dimensions, color, taste, smell, or feel of their constituent units. Not all food has
an attractive smell, for example. Beds, chairs, chests of drawers, and tables need not fall in
a unique range of shapes. So once we have a sufficient body of concepts drawn directly
from perception, we can employ conceptual distinctions, such as functions, in the CCDs we
use to form new concepts.
The process of conceptual identification (of subsuming a new concrete under an
appropriate concept) is learned as one learns to speak, and it becomes automatic in
the case of existents given in perceptual awareness. . . . But it becomes
progressively more difficult as mans concepts move farther away from direct

7
perceptual evidence, and involve complex combinations and cross-classifications of
many early concepts. (Observe the difficulties of identifying a given political
system or of diagnosing a rare disease.) In such cases, the knowledge of whether a
concrete is or is not to be subsumed under a certain concept does not come
automatically, but requires a new cognitive effort.

We speak of an abstract concept as one that is very wide. But in Rands theory, the
measure of abstractness is the number of abstractions required to form and employ the
concept, which is the measure of the effort required to really understand what the concept
means.

Varieties of Measurement
Although Rand uses cardinal measurement as her paradigm examples (length and
table), she intends the term measurement to apply to a wide variety of means of
objectively comparing two or more existents. Mental states, for example, may be measured
by the scope of factual material involved in a given cognitive process and by the length of
the conceptual chain required to deal with that material. (32) She states that concepts
pertaining to evaluation maybe be measured ordinally, in a process she terms teleological
measurement, which grades or ranks its objects in terms of the degree to which they
achieve or frustrate some goal or end. (33)
Rand also mentions concepts of method, which are a sub-category of concepts
pertaining to the products of consciousness. Concepts of method designate systematic
courses of action devised by men for the purpose of achieving certain goals. (35) She says
that Concepts of method represent a large part of mans conceptual equipment. (36). She
gives as examples logic and many ideas in the sciences, ethics, medicine, etc., and in the
appendix analyzes complex numbers as concepts of method. (305-6)
The concepts of method are the link to the vast and complex category of concepts that
represent integrations of existential concepts with concepts of consciousness. (36) These
include concepts such as marriage and property, which refer to human relations that
can only be understood in context, via a complex conceptual chain.
Thus Rands theory holds that all conceptual categories have a basis in the range of
measurements omitted on one or more Conceptual Common Denominators. In the
paradigm cases, this process is based entirely on the data of the senses. But many concepts,
and especially concepts of human relations, methods, and knowledge are based on complex
chains of abstraction in which abstractions themselves play a part in the CCDs. This is not a
circular position as long as all the abstractions employed in the CCDs have themselves a
basis that is ultimately rooted in perception.

8
Definitions and Essentials
In her initial definition of a concept, Rand stated that a concept is united by a specific
definition. According to Rand: The purpose of a definition is to distinguish a concept
from all other concepts and thus to keep its units differentiated from all other existents.
(40)
Since the definition of a concept is formulated in terms of other concepts, it
enables man, not only to identify and retain a concept, but also to establish
the relationships, the hierarchy, the integration of all his concepts and thus
the integration of his knowledge. Definitions preserve, not the chronological
order in which a given man may have learned concepts, but the logical order
of their hierarchical interdependence. (40)
Rand endorses genus-differentia definitions, and describes how this system relates to the
main elements of her theory of abstraction:
The units of a concept were differentiatedby means of a distinguishing
characteristic(s)from other existents possessing a commensurable
characteristic, a Conceptual Common Denominator. A definition follows
the same principle: it specifies the distinguishing characteristic(s) of the
units, and indicates the category of existents from which they were
differentiated.
The distinguishing characteristic(s) of the units becomes the
differentia of the concepts definition; the existents possessing a
Conceptual Common Denominator become the genus. (41)
To illustrate this view of definitions, Rand discusses the concept man, defined as a
rational animal. She points out that animal is the genus of man because mans
distinguishing characteristic is reason; and only animals possess consciousness, the CCD
along which we differentiate and integrate people into man.
This approach has three crucial implications for the nature of definitions:
1) Definitions are not the meaning of a concept. A definition is not a description; it
implies, but does not mention all the characteristics of a concepts units. . . . A definition
must identify the nature of the units, i.e. the essential characteristics without which the
units would not be the kind of existents they are. (42)
2) Concepts are contextual. Concepts are not and cannot be formed in a vacuum; they are
formed in context; the process of conceptualization consists of observing the differences
and similarities of the existents within the field of ones awareness. . . . From a childs
grasp of the simplest concept integrating a group of perceptually given concretes, to a
scientists grasp of the most complex abstractions integrating long conceptual chains. . .
the context is the entire field of a minds awareness or knowledge. . . .
3) Definitions can be true or false. A Definition is the condensation of a vast body of

9
observationsand stands or falls with the truth or falsehood of these observations. (48)
Thus concepts are formed in relation to the needs and capacities of an individual, based in
his context of knowledge, and constrained by the real similarities and differences that can
be discerned along dimensions of comparison in reality.
This approach allows Rand to answer the question of what an essential property is.
Having denied that there are any metaphysical universals or essences (52), Rand
reconceives of essentials in terms of fundamental characteristics.
Metaphysically, a fundamental characteristic [of an existent] is that
distinctive characteristic which makes the greatest number of others
possible; epistemologically, it is the one that explains the greatest number of
others. (45)
Thus what is essential depends in part on what it is that one trying to understand or explain.
And it depends on what is in fact the nature of the existents under consideration. As Rand
says, in all issues pertaining to objectivity, there is no ultimate authority except reality.
In this way Rand conceives of a view of conceptsand of knowledge more generallythat
regards human beings as active but objective knowers. Reality constrains and informs
human knowledge, and it must be complied with for concepts to be objective; but human
choice and human purposes are also relevant to understanding why a concept might be
necessary in the first place, and to understanding how it might be grasped and defined.
Rands theory gives a new sense to the idea of objectivity, allowing her to fulfill her
intention to distinguish her view from realism (whose epistemological tendency she dubs
intrinsicism) and nominalism (whose tendency she dubs subjectivism). (534)

Sources:

Ayn Rand. Introduction to Objectivist Epistemology, Expanded Second Edition, Harry

Binswanger and Leonard Peikoff eds. (New York: Meridian, 1990).
(First edition originally published in The Objectivist July 1966-February 1967)

A useful follow-up is David Kelley. A Theory of Abstraction Cognition and Brain

Theory, 1984, 7 (3 & 4), pp. 329357 (reprinted in pamphlet form by The Objectivist
Center). Kelley extends Rands theory and relates it to research in psychology and
neuroscience.

10
Appendix A:
The is a partial listing of some of the Conceptual Common Denominators Rand proposes
(many of these are implicit in her presentations of definitions or distinguishing
characteristics. Where I could not clearly determine to my satisfaction what the CCD might
be, I have included the distinguishing characteristics Rand mentions.
Concept

Conceptual Common Denominator(s)

Page

Table

Shape, number of supports, size

12

Furniture
Animal

Size, ability to perform functions (such as support human 22

beings or store smaller items)
Locomotion and range of consciousness
24

Fish Amphibian
Bird Mammal

CCD of animal qualified by anatomical and

physiological characteristics

Man (philosophical
context)

CCD of animal plus degree of cognitive ability (range

of consciousness)

25

Man(infants
context)

Locomotion and speech

43

Man (toddlers
context)

Number of legs, degree of furriness, locomotion, and

speech.

44

Man (childs
context)
adult child
adolescent
American,
Englishman etc.

CCD of living thing and quantity and scale of unique

activities.
CCD of man plus measurements of age

44

CCD of man plus national origin

26

Thought

Degree, intensity, and directedness psychological

action Distinguishing characteristic: a purposefully
directed process of cognition.
Degree of affection. Distinguishing characteristics: an
emotion proceeding from the evaluation of an existent as
a positive value and as a source of pleasure.
Degree of achievement or frustration of a standard of
value.

32

Love

Good or other
moral term.

25

25

34

33